Reduce some steps

.gitignore

@@ -0,0 +1 @@
+.venv

.python-version

@@ -0,0 +1 @@
+3.10

app.py

@@ -6,88 +6,67 @@ import os
 from typing import Optional, Dict, Any, List, Tuple
 
 # Available OuteTTS models based on the documentation
-MODELS = {v.value: k for k, v in outetts.Models.__members__.items()}
+MODELS = {v.value: v for _, v in outetts.Models.__members__.items()}
 
-def initialize_interface(model_name: str) -> Tuple[str, str, Any]:
-    """Initialize the OuteTTS interface with selected model and backend."""
-    try:
-        model = MODELS[model_name]
-        
-        # Configure the model
-        config = outetts.ModelConfig.auto_config(
-            model=model,
-            backend=outetts.Backend.LLAMACPP,
-            quantization=outetts.LlamaCppQuantization.FP16,
-        )
-        
-        # Initialize the interface
-        interface = outetts.Interface(config=config)
-        
-        status = f"✅ Successfully initialized {model_name}"
-        info = f"Model: {model_name}\nStatus: Ready"
-        
-        return status, info, interface
-        
-    except Exception as e:
-        error_msg = f"❌ Error initializing model: {str(e)}"
-        return error_msg, f"Error: {str(e)}", None
+def initialize_interface(model_name: str) -> Any:
+    """Initialize the OuteTTS interface with selected model."""
+    model = MODELS[model_name]
+    
+    # Configure the model
+    config = outetts.ModelConfig.auto_config(
+        model=model,
+        backend=outetts.Backend.LLAMACPP,
+        quantization=outetts.LlamaCppQuantization.FP16,
+    )
+    
+    # Initialize the interface
+    interface = outetts.Interface(config=config)
+    return interface
 
-def create_speaker_from_audio(audio_file, interface_state) -> Tuple[str, str, Any]:
+def create_speaker_from_audio(audio_file, interface_state) -> Tuple[str, Any]:
     """Create a speaker profile from uploaded audio file."""
     if interface_state is None:
-        return "❌ Please initialize a model first", "", None
+        raise gr.Error("Please select a model first")
     
     if audio_file is None:
-        return "❌ Please upload an audio file", "", None
-    
-    try:
-        # Create speaker profile from audio
-        speaker = interface_state.create_speaker(audio_file)
-        
-        # Convert speaker dict to formatted JSON
-        speaker_json = json.dumps(speaker, indent=2, ensure_ascii=False)
-        
-        status = "✅ Speaker profile created successfully!"
-        
-        return status, speaker_json, speaker
-        
-    except Exception as e:
-        error_msg = f"❌ Error creating speaker: {str(e)}"
-        return error_msg, f"Error: {str(e)}", None
+        raise gr.Error("Please upload an audio file")
+    
+    # Create speaker profile from audio
+    speaker = interface_state.create_speaker(audio_file)
+    
+    # Convert speaker dict to formatted JSON
+    speaker_json = json.dumps(speaker, indent=2, ensure_ascii=False)
+    
+    return speaker_json, speaker
 
 
 
-def generate_sample_audio(text: str, temperature: float, interface_state, speaker_state) -> Tuple[str, Optional[str]]:
+def generate_sample_audio(text: str, temperature: float, interface_state, speaker_state) -> Optional[str]:
     """Generate sample audio using the current speaker."""
     if interface_state is None:
-        return "❌ Please initialize a model first", None
+        raise gr.Error("Please select a model first")
     
     if speaker_state is None:
-        return "❌ Please create a speaker first", None
+        raise gr.Error("Please create a speaker first")
     
     if not text.strip():
-        return "❌ Please enter text to generate", None
-    
-    try:
-        # Generate audio
-        output = interface_state.generate(
-            config=outetts.GenerationConfig(
-                text=text,
-                speaker=speaker_state,
-                sampler_config=outetts.SamplerConfig(
-                    temperature=temperature
-                )
+        raise gr.Error("Please enter text to generate")
+    
+    # Generate audio
+    output = interface_state.generate(
+        config=outetts.GenerationConfig(
+            text=text,
+            speaker=speaker_state,
+            sampler_config=outetts.SamplerConfig(
+                temperature=temperature
             )
         )
-        
-        # Save to temporary file
-        with tempfile.NamedTemporaryFile(suffix=".wav", delete=False) as f:
-            output.save(f.name)
-            return "✅ Audio generated successfully!", f.name
-            
-    except Exception as e:
-        error_msg = f"❌ Error generating audio: {str(e)}"
-        return error_msg, None
+    )
+    
+    # Save to temporary file
+    with tempfile.NamedTemporaryFile(suffix=".wav", delete=False) as f:
+        output.save(f.name)
+        return f.name
 
 # Create the Gradio interface
 with gr.Blocks(title="OuteTTS Speaker Creator", theme=gr.themes.Soft()) as demo:
@@ -100,19 +79,15 @@ with gr.Blocks(title="OuteTTS Speaker Creator", theme=gr.themes.Soft()) as demo:
     
     with gr.Row():
         with gr.Column(scale=1):
-            gr.Markdown("## 🔧 Model Configuration")
+            gr.Markdown("## 🔧 Model Selection")
             
             model_dropdown = gr.Dropdown(
                 choices=list(MODELS.keys()),
-                value=list(MODELS.keys())[0],
+                value=list(MODELS.keys())[-1],  # Default to last model
                 label="Select OuteTTS Model",
                 info="Choose the model variant to use"
             )
             
-            init_button = gr.Button("Initialize Model", variant="primary")
-            init_status = gr.Textbox(label="Status", interactive=False)
-            init_info = gr.Textbox(label="Model Info", interactive=False, lines=3)
-            
         with gr.Column(scale=1):
             gr.Markdown("## 🎵 Speaker Creation")
             
@@ -122,8 +97,6 @@ with gr.Blocks(title="OuteTTS Speaker Creator", theme=gr.themes.Soft()) as demo:
                 sources=["upload", "microphone"]
             )
             create_button = gr.Button("Create Speaker Profile", variant="primary")
-            
-            speaker_status = gr.Textbox(label="Speaker Status", interactive=False)
     
     with gr.Row():
         with gr.Column():
@@ -156,26 +129,32 @@ with gr.Blocks(title="OuteTTS Speaker Creator", theme=gr.themes.Soft()) as demo:
             )
             
             generate_button = gr.Button("Generate Sample Audio", variant="primary")
-            generate_status = gr.Textbox(label="Generation Status", interactive=False)
             sample_audio = gr.Audio(label="Generated Audio", interactive=False)
     
     # Event handlers
-    init_button.click(
+    # Initialize default model on page load
+    demo.load(
+        fn=initialize_interface,
+        inputs=[model_dropdown],
+        outputs=[interface_state]
+    )
+    
+    model_dropdown.change(
         fn=initialize_interface,
         inputs=[model_dropdown],
-        outputs=[init_status, init_info, interface_state]
+        outputs=[interface_state]
     )
     
     create_button.click(
         fn=create_speaker_from_audio,
         inputs=[audio_upload, interface_state],
-        outputs=[speaker_status, speaker_json, speaker_state]
+        outputs=[speaker_json, speaker_state]
     )
     
     generate_button.click(
         fn=generate_sample_audio,
         inputs=[test_text, temperature_slider, interface_state, speaker_state],
-        outputs=[generate_status, sample_audio]
+        outputs=[sample_audio]
     )
 
 demo.launch()

requirements.txt

@@ -1,3 +1,2 @@
-llama-cpp-python --extra-index-url https://abetlen.github.io/llama-cpp-python/whl/cpu
 outetts==0.4.4
 gradio